In industrial machines including molding machines such as injection molding machines and press molding machines as well as bonding machines, pressure control has been adopted. With pressure control, a pressure signal from a machine is detected when a mechanical drive unit driven by a motor is pressed against a workpiece and the like, which act as pressurization targets; a pressure control computation is set by a control parameter on the basis of the pressure signal; and target pressure is applied, so that the motor is operated. The control parameter is the gain and the like of the pressure control computation.
It is necessary to appropriately adjust the control parameter during the pressure control described above. If a control parameter such as gain is excessively large, oscillation occurs and thus the control system operates unstably. Or, even if the operation of the control system does not become unstable, a small vibration is generated in a pressure signal and it is transmitted to the workpiece and the like, and the result of the machining is adversely affected. Conversely, if the control parameter is excessively small, it takes time to reach the target pressure (the command signal for the pressure, the pressure command signal). Further, when disturbance is applied, this disturbance cannot be removed completely. Particularly, the disturbance cannot be compensated for only by feedforward control and is only removed by the pressure control computation (feedback control) performed on the basis of a deviation of a pressure signal from the target pressure. Accordingly, it is important to appropriately adjust a control parameter of a pressure control unit.
For example, Patent Literature 1 discloses a pressure control technique that involves multiplying the pressure deviation of a pressure signal away from the target pressure with a pressure gain so as to determine a motor speed command and by performing a speed control computation that follows the speed command, in which an elastic constant of a pressurization target is calculated and then divided by a predetermined proportional constant so that a pressure gain is calculated.